Variation, selection and evolution of function-valued traits.

Area of Science:

• Evolutionary Biology
• Quantitative Genetics
• Mathematical Modeling

Background:

• Understanding phenotypic trait evolution typically uses multivariate models for quantitative traits.
• Function-valued traits, which are mathematical functions, present unique challenges and require advanced modeling approaches.
• Existing models may not fully capture the constraints on variation inherent in continuous, function-valued traits.

Purpose of the Study:

• To present a novel framework for analyzing the variation, selection, and evolution of function-valued phenotypic traits.
• To demonstrate the application of this framework using thermal performance curves (TPCs) of caterpillar growth rates.
• To extend existing quantitative genetic models to accommodate function-valued traits.

Main Methods:

• Quantified phenotypic and genetic variation using variance-covariance functions and eigenfunctions for TPCs.
• Defined selection on function-valued traits using selection gradient functions, relating performance to fitness across environmental conditions.
• Predicted evolutionary responses by integrating genetic variation and selection gradient functions.

Main Results:

• Demonstrated that function-valued traits have unique variation patterns not captured by standard multivariate models.
• Showed that selection gradients for TPCs are influenced by environmental state distributions (e.g., temperatures).
• Predicted potentially non-linear evolutionary responses of TPCs, even with linear mean phenotypes and selection gradients.

Conclusions:

• The proposed framework offers a robust method for studying the evolution of function-valued traits.
• Function-valued traits exhibit complex variation and selection dynamics that necessitate specialized modeling.
• Future research should address methodological and empirical challenges in studying the evolution of these traits.